FOTD -- May 22, 1016 (Rating A-7,M-7) Fractal visionaries and enthusiasts: Today's image shows a scene in the parent fractal that results when Z^(2) and Z^(-1) are blended. Since there is no such thing as a minibrot of a negative unity order, (it would have negative-two main bays), all minibrots reach a floor at the quadratic level, with one lobe. Today's minibrot did its best to be impressive, but I had to add some Fractint special effects to bring it up to FOTX worthiness. Rendering the image with the inside set to 'bof60' and the outside set to 'fmod' gave it just enough bells and whistles to make it worth being declared FOTX for May 22. The reason for the name "MothEaten Minibrot" will become clear when the three minute parameter file is run in Fractint. The next FOTD, which very possibly might be posted in two days or sooner, will feature a fractal between quadratic and cubic that has no discontinuities and has enough zooms to make the most dedicated fan of zooming leap for joy. It rained almost all day here at Fractal Central today, forcing myself and FL to spend the day indoors with the fractal cats. Until next time, take care, and trying to squeeze the quantum world into the world of classic Newtonian common sense is as futile as trying to squeeze a 4-dimensional hypercube into the 3-dimensional world without distorting it. Yet this is what the different interpretations of QM are apparently trying to do. Jim Muth jimmuth@earthlink.net START PARAMETER FILE======================================= MothEaten_Minibrot { ; time=0:03:14.60 SF5 at 200MHZ reset=2004 type=formula formulafile=basicer.frm formulaname=FinDivBrot-2 function=recip passes=1 center-mag=-0.219974943536/+1.166215418628/775/1/\ 4.75/0 params=-1/6.25/10/0 float=y maxiter=1500 inside=bof60 outside=fmod logmap=yes periodicity=6 colors=000000gRP`XQUcQNiRGpSzzSGUL1JE7EFD6ZJ4hO2rW\ CjcLbkUWzcclWSlWQgVNbUKZTHUSFPRCLQ9GP6CP4AUA8ZF6cK\ 4hQ2mV1r_5la8gcCbeFYfIShMNjPIlSDm_IafNRgKPhINiFLzT\ JjBHgEGdGGaIG_KGXMGUOGSQGPTFMVFJXFHZFE`FBbF9dFGeKM\ ePSfUYfZcfbbe`ae_`eZ_eYZeXYeWXdVWdUVdTUdSTdRSdQRdP\ QcOPcNOcMNcLMcKLcJLcIkeHgfGchG_iFWjFSkFQ`RPQaOFlN4\ wM9tLDqLHnKLkKQhJUeJYcIa`HeYHjVGnSGrPFvMFzKHsIImGJ\ gFLaDtWCzPAPJ8QD7R75S14R26z37Rz8Q5AQ6BQ7Cz8DUAMYCU\ aEbeGjZJhSMfLOdERb8Ta9VW9WR9YL9ZGA_FB_EC`ED`DDaDEa\ CFbBGbBHcAHcAId9Jd8Ke8Le7Le7LfALfDLgGLgJLhLLhOLiRL\ iULjXLjZLkaLkdLlgLljLmlLmoLnrLnuLoxLozOnxQmwSlvVku\ XjtZirahzcgzefzgezjdzlcznbzqazs`zu_zwZzuYzsYzqYzoY\ zmYzkYziYcgYbeYbcYbaYaNEsLEsKEsJEsIEsHEqGEoGEmFEkF\ EiEEhEEfDEdDEbCE`CE_DF`EFaEGbFGcGHdGHdHIeIIfIJgJJh\ KKiKKiMLhNLgOMfPMeQNdSNcTObUOaVP`WP_YQZZQY_RX`RWaR\ VTQTKQSBQR2QQ3NR3LS4JTnKO } frm:FinDivBrot-2 { ; Jim Muth z=(0,0), c=pixel, a=-(real(p1)-2), esc=(real(p2)+16), b=imag(p1): z=(b)*(z*z*fn1(z^(a)+b))+c |z| < esc } END PARAMETER FILE=========================================
The late Paul N. Lee suggested I try the image processing program GIMP. I had not tried GIMP until my anti-aliasing results in Photoshop (using their best down-sampling algorithm -- "Bi-cubic sharper") on this Win 8.1 computer came out poorer quality than the Lanczos algorithm used in my not-supported (on Win 8.1) image processing program I had used on my Win-XP computer: "CompuPic". GIMP *has* the Lanczos algorithm, and I have used it on this set of my investigations of Jim's FOTX -- May 22, 2016 image. I followed the Lanczos down-sampling with a small amount of sharpening to attempt to remove some of the "softness" of edges created by that algorithm. I like the results much better. I'm not yet familiar with GIMP and inadvertently left the default "Progressive" checkbox checked, so the JPG images appear in three "waves" of increasing resolution. I'll try and remember to uncheck that box for the next set of images. As usual, thanks for viewing the images full size. Jim's FOTX -- May 22, 2016: http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522.jpg All the following images have LogMap changed from 1 to 0 to try to get as many colors of Jim's color map used as possible in these images. The FOTX's "parent": - 4X oversampled http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522Z.jpg My investigations into the parent: - 4X oversampled http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522Y.jpg The next two pairs of images are of the same areas, but one version is not anti-aliased and the other is. To rapidly compare them, try putting an image pair in adjacent browser tabs towards the left side of your row of tabs. Then alternate typing these key combinations (if they're in the 2nd & 3rd tabs, for example): <ctrl><2> and <ctrl><3> This works in most browsers. - JPG is 4X oversampled http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522X.gif http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522W.jpg Here's another similar pair: - MaxIter changed from 1500 to 3000 for both - JPG is 5X oversampled http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522U.gif http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522T.jpg And a zoom into the heart of F160522T.jpg: - MaxIter changed from 1500 to 3000 - 5X oversampled http://www.emarketingiseasy.com/TESTS/FOTD/2016/F160522S.jpg - Hal Lane ######################## # hallane@earthlink.net ######################## -----Original Message----- From: Fractint [mailto:fractint-bounces@mailman.xmission.com] On Behalf Of Jim Muth Sent: Sunday, May 22, 2016 8:21 PM To: fractint <fractint@mailman.xmission.com> Subject: [Fractint] FOTX -- May 22, 2016 FOTD -- May 22, 1016 (Rating A-7,M-7) Fractal visionaries and enthusiasts: Today's image shows a scene in the parent fractal that results when Z^(2) and Z^(-1) are blended. Since there is no such thing as a minibrot of a negative unity order, (it would have negative-two main bays), all minibrots reach a floor at the quadratic level, with one lobe. Today's minibrot did its best to be impressive, but I had to add some Fractint special effects to bring it up to FOTX worthiness. Rendering the image with the inside set to 'bof60' and the outside set to 'fmod' gave it just enough bells and whistles to make it worth being declared FOTX for May 22. The reason for the name "MothEaten Minibrot" will become clear when the three minute parameter file is run in Fractint. The next FOTD, which very possibly might be posted in two days or sooner, will feature a fractal between quadratic and cubic that has no discontinuities and has enough zooms to make the most dedicated fan of zooming leap for joy. It rained almost all day here at Fractal Central today, forcing myself and FL to spend the day indoors with the fractal cats. Until next time, take care, and trying to squeeze the quantum world into the world of classic Newtonian common sense is as futile as trying to squeeze a 4-dimensional hypercube into the 3-dimensional world without distorting it. Yet this is what the different interpretations of QM are apparently trying to do. Jim Muth jimmuth@earthlink.net START PARAMETER FILE======================================= MothEaten_Minibrot { ; time=0:03:14.60 SF5 at 200MHZ reset=2004 type=formula formulafile=basicer.frm formulaname=FinDivBrot-2 function=recip passes=1 center-mag=-0.219974943536/+1.166215418628/775/1/\ 4.75/0 params=-1/6.25/10/0 float=y maxiter=1500 inside=bof60 outside=fmod logmap=yes periodicity=6 colors=000000gRP`XQUcQNiRGpSzzSGUL1JE7EFD6ZJ4hO2rW\ CjcLbkUWzcclWSlWQgVNbUKZTHUSFPRCLQ9GP6CP4AUA8ZF6cK\ 4hQ2mV1r_5la8gcCbeFYfIShMNjPIlSDm_IafNRgKPhINiFLzT\ JjBHgEGdGGaIG_KGXMGUOGSQGPTFMVFJXFHZFE`FBbF9dFGeKM\ ePSfUYfZcfbbe`ae_`eZ_eYZeXYeWXdVWdUVdTUdSTdRSdQRdP\ QcOPcNOcMNcLMcKLcJLcIkeHgfGchG_iFWjFSkFQ`RPQaOFlN4\ wM9tLDqLHnKLkKQhJUeJYcIa`HeYHjVGnSGrPFvMFzKHsIImGJ\ gFLaDtWCzPAPJ8QD7R75S14R26z37Rz8Q5AQ6BQ7Cz8DUAMYCU\ aEbeGjZJhSMfLOdERb8Ta9VW9WR9YL9ZGA_FB_EC`ED`DDaDEa\ CFbBGbBHcAHcAId9Jd8Ke8Le7Le7LfALfDLgGLgJLhLLhOLiRL\ iULjXLjZLkaLkdLlgLljLmlLmoLnrLnuLoxLozOnxQmwSlvVku\ XjtZirahzcgzefzgezjdzlcznbzqazs`zu_zwZzuYzsYzqYzoY\ zmYzkYziYcgYbeYbcYbaYaNEsLEsKEsJEsIEsHEqGEoGEmFEkF\ EiEEhEEfDEdDEbCE`CE_DF`EFaEGbFGcGHdGHdHIeIIfIJgJJh\ KKiKKiMLhNLgOMfPMeQNdSNcTObUOaVP`WP_YQZZQY_RX`RWaR\ VTQTKQSBQR2QQ3NR3LS4JTnKO } frm:FinDivBrot-2 { ; Jim Muth z=(0,0), c=pixel, a=-(real(p1)-2), esc=(real(p2)+16), b=imag(p1): z=(b)*(z*z*fn1(z^(a)+b))+c |z| < esc } END PARAMETER FILE========================================= _______________________________________________ Fractint mailing list Fractint@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/fractint --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
I use GIMP all the time, being of sound Linux mind ;). And the Lanczos setting for image resizes. Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. You might like the results. Thanks for the images, now just need to start rendering them at full Fractint max size, downsample to 25%, and print posters! On 05/23/2016 04:33 PM, Harold Lane wrote:
The late Paul N. Lee suggested I try the image processing program GIMP. I had not tried GIMP until my anti-aliasing results in Photoshop (using their best down-sampling algorithm -- "Bi-cubic sharper") on this Win 8.1 computer came out poorer quality than the Lanczos algorithm used in my not-supported (on Win 8.1) image processing program I had used on my Win-XP computer: "CompuPic".
GIMP *has* the Lanczos algorithm, and I have used it on this set of my investigations of Jim's FOTX -- May 22, 2016 image. I followed the Lanczos down-sampling with a small amount of sharpening to attempt to remove some of the "softness" of edges created by that algorithm.
I like the results much better.
I'm not yet familiar with GIMP and inadvertently left the default "Progressive" checkbox checked, so the JPG images appear in three "waves" of increasing resolution. I'll try and remember to uncheck that box for the next set of images.
-- David W. Jones gnome@hawaii.rr.com authenticity, honesty, community http://dancingtreefrog.com
----- David said: --------------
Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. I saw a similar function in Photoshop, but I don't know how the three controls interact -- or even what each control's effect is supposed to be... The default settings (right after GIMP's installation) seem to alter fractal images a bit too much.
So, when features in an image occasionally appear to need a touch of sharpening, I've been using low (e. g. 20) settings of the single control: Filters > Enhance > Sharpen... I guess it's a filter with training wheels for noobs. ;P But, amazingly, many of the Lanczos down-sampled images don't really appear to need any sharpening at all -- especially if I render them at 5X the final image size. Can you tell me (or point me to a page) how the three Unsharp Mask controls work? I read the GIMP Help, but I'm not sure if the assumptions and procedures for sharpening real world photography hold for our fractals... - Hal Lane ######################## # hallane@earthlink.net ######################## -----Original Message----- From: Fractint [mailto:fractint-bounces@mailman.xmission.com] On Behalf Of david Sent: Tuesday, May 24, 2016 4:35 AM To: Fractint and General Fractals Discussion <fractint@mailman.xmission.com> Subject: Re: [Fractint] FOTX -- May 22, 2016 I use GIMP all the time, being of sound Linux mind ;). And the Lanczos setting for image resizes. Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. You might like the results. Thanks for the images, now just need to start rendering them at full Fractint max size, downsample to 25%, and print posters! On 05/23/2016 04:33 PM, Harold Lane wrote:
The late Paul N. Lee suggested I try the image processing program GIMP. I had not tried GIMP until my anti-aliasing results in Photoshop (using their best down-sampling algorithm -- "Bi-cubic sharper") on this Win 8.1 computer came out poorer quality than the Lanczos algorithm used in my not-supported (on Win 8.1) image processing program I had used on my Win-XP computer: "CompuPic".
GIMP *has* the Lanczos algorithm, and I have used it on this set of my investigations of Jim's FOTX -- May 22, 2016 image. I followed the Lanczos down-sampling with a small amount of sharpening to attempt to remove some of the "softness" of edges created by that algorithm.
I like the results much better.
I'm not yet familiar with GIMP and inadvertently left the default "Progressive" checkbox checked, so the JPG images appear in three "waves" of increasing resolution. I'll try and remember to uncheck that box for the next set of images.
-- David W. Jones gnome@hawaii.rr.com authenticity, honesty, community http://dancingtreefrog.com _______________________________________________ Fractint mailing list Fractint@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/fractint --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
I'm not practiced with using the Unsharp Mask on fractals, but the parameter I usually adjust the most is the first one, Radius, for photos. But reading the GIMP help was interesting, including what I think the Unsharp Mask is doing. First, they recommend converting the image to Hue, Saturation and Value layers, then using Unsharp mask only on the Value layer. I never knew that! When done, convert it back to RGB format. Anyway, first the Unsharp Mask filter creates two duplicate layers in the image. It blurs one of them, then merges it with the original image by *subtracting* the value of each pixel from its value in the original image. Then it merges the second duplicate layer into the original image (as modified by the above subtract process) by *adding the value of each pixel in it to the value of the corresponding pixel in the original. So basically it produces extra sharpening along edges without blurring areas without edges. A mild sharpening like you're using is almost unnoticeable on the image I used. Now, to me, most Fractint GIFs are edges containing areas of flat color. That is unlike almost any photograph; very few photos have areas of identical color values - while they may look quite close or similar, chances are that neighboring pixels in that area will have slightly different values. So all I can suggest is take advantage of the Preview toggle on GIMP's Unsharp Mask filter. Change a setting, toggle Preview on/off to see the effect. You can view different parts of the image in the Preview window, too, to see how it affects areas with lots of details and areas with few details. About the settings: Radius determines how many pixels on each side of an edge are affected by the sharpening. It looks to me like the larger the radius, the wider the area of changed color is. Try giving it a large value to make the effect more noticable and toggle Preview on/off to see the difference. To me, the default radius of 5 seemed to bring out more detail in the areas with fine detail. Changing it to 10 seemed to darken a number of colors more, setting it to 20 even more. Amount determines how much color change to make. Try a large value like 10 and toggle Preview on/off to see the difference. On a smaller fractal I tried it on, a value of 10 looked like it was almost adding more bands to the image. That may not be something you want on a fractal! Threshold tells the filter how much of a color difference you want it to consider an "edge". Its values don't seem to have much noticeable effect on the fractal I just fiddled around with. Just based on the general recommendation from digital photography folk, I'd say using Unsharp Mask anytime you'd use Sharpening. But you could try each one on your downsampled images and see which you like best. Now I need to go generate myself a big Fractint image and see what effect Unsharp Mask has on downsampled images ... On 05/23/2016 11:51 PM, Harold Lane wrote:
----- David said: --------------
Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. I saw a similar function in Photoshop, but I don't know how the three controls interact -- or even what each control's effect is supposed to be... The default settings (right after GIMP's installation) seem to alter fractal images a bit too much.
So, when features in an image occasionally appear to need a touch of sharpening, I've been using low (e. g. 20) settings of the single control: Filters > Enhance > Sharpen...
I guess it's a filter with training wheels for noobs. ;P
But, amazingly, many of the Lanczos down-sampled images don't really appear to need any sharpening at all -- especially if I render them at 5X the final image size.
Can you tell me (or point me to a page) how the three Unsharp Mask controls work? I read the GIMP Help, but I'm not sure if the assumptions and procedures for sharpening real world photography hold for our fractals...
- Hal Lane
######################## # hallane@earthlink.net ########################
-----Original Message----- From: Fractint [mailto:fractint-bounces@mailman.xmission.com] On Behalf Of david Sent: Tuesday, May 24, 2016 4:35 AM To: Fractint and General Fractals Discussion <fractint@mailman.xmission.com> Subject: Re: [Fractint] FOTX -- May 22, 2016
I use GIMP all the time, being of sound Linux mind ;). And the Lanczos setting for image resizes.
Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. You might like the results.
Thanks for the images, now just need to start rendering them at full Fractint max size, downsample to 25%, and print posters!
On 05/23/2016 04:33 PM, Harold Lane wrote:
The late Paul N. Lee suggested I try the image processing program GIMP. I had not tried GIMP until my anti-aliasing results in Photoshop (using their best down-sampling algorithm -- "Bi-cubic sharper") on this Win 8.1 computer came out poorer quality than the Lanczos algorithm used in my not-supported (on Win 8.1) image processing program I had used on my Win-XP computer: "CompuPic".
GIMP *has* the Lanczos algorithm, and I have used it on this set of my investigations of Jim's FOTX -- May 22, 2016 image. I followed the Lanczos down-sampling with a small amount of sharpening to attempt to remove some of the "softness" of edges created by that algorithm.
I like the results much better.
I'm not yet familiar with GIMP and inadvertently left the default "Progressive" checkbox checked, so the JPG images appear in three "waves" of increasing resolution. I'll try and remember to uncheck that box for the next set of images.
-- David W. Jones gnome@hawaii.rr.com authenticity, honesty, community http://dancingtreefrog.com
David, Thank you for the detailed explanation of your experiences with unsharp masking. Your info is valuable & I propose to carefully reread and try it out on the batch of images I create from Jim's next FOTD. Thanks again for the detailed info -- and the analysis of how the layer subtraction process differs in large single color areas from that at edges! - Hal Lane ######################## # hallane@earthlink.net ######################## -----Original Message----- From: Fractint [mailto:fractint-bounces@mailman.xmission.com] On Behalf Of david Sent: Wednesday, May 25, 2016 3:13 AM To: Fractint and General Fractals Discussion <fractint@mailman.xmission.com> Subject: Re: [Fractint] FOTX -- May 22, 2016 I'm not practiced with using the Unsharp Mask on fractals, but the parameter I usually adjust the most is the first one, Radius, for photos. But reading the GIMP help was interesting, including what I think the Unsharp Mask is doing. First, they recommend converting the image to Hue, Saturation and Value layers, then using Unsharp mask only on the Value layer. I never knew that! When done, convert it back to RGB format. Anyway, first the Unsharp Mask filter creates two duplicate layers in the image. It blurs one of them, then merges it with the original image by *subtracting* the value of each pixel from its value in the original image. Then it merges the second duplicate layer into the original image (as modified by the above subtract process) by *adding the value of each pixel in it to the value of the corresponding pixel in the original. So basically it produces extra sharpening along edges without blurring areas without edges. A mild sharpening like you're using is almost unnoticeable on the image I used. Now, to me, most Fractint GIFs are edges containing areas of flat color. That is unlike almost any photograph; very few photos have areas of identical color values - while they may look quite close or similar, chances are that neighboring pixels in that area will have slightly different values. So all I can suggest is take advantage of the Preview toggle on GIMP's Unsharp Mask filter. Change a setting, toggle Preview on/off to see the effect. You can view different parts of the image in the Preview window, too, to see how it affects areas with lots of details and areas with few details. About the settings: Radius determines how many pixels on each side of an edge are affected by the sharpening. It looks to me like the larger the radius, the wider the area of changed color is. Try giving it a large value to make the effect more noticable and toggle Preview on/off to see the difference. To me, the default radius of 5 seemed to bring out more detail in the areas with fine detail. Changing it to 10 seemed to darken a number of colors more, setting it to 20 even more. Amount determines how much color change to make. Try a large value like 10 and toggle Preview on/off to see the difference. On a smaller fractal I tried it on, a value of 10 looked like it was almost adding more bands to the image. That may not be something you want on a fractal! Threshold tells the filter how much of a color difference you want it to consider an "edge". Its values don't seem to have much noticeable effect on the fractal I just fiddled around with. Just based on the general recommendation from digital photography folk, I'd say using Unsharp Mask anytime you'd use Sharpening. But you could try each one on your downsampled images and see which you like best. Now I need to go generate myself a big Fractint image and see what effect Unsharp Mask has on downsampled images ... On 05/23/2016 11:51 PM, Harold Lane wrote:
----- David said: --------------
Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. I saw a similar function in Photoshop, but I don't know how the three controls interact -- or even what each control's effect is supposed to be... The default settings (right after GIMP's installation) seem to alter fractal images a bit too much.
So, when features in an image occasionally appear to need a touch of sharpening, I've been using low (e. g. 20) settings of the single control: Filters > Enhance > Sharpen...
I guess it's a filter with training wheels for noobs. ;P
But, amazingly, many of the Lanczos down-sampled images don't really appear to need any sharpening at all -- especially if I render them at 5X the final image size.
Can you tell me (or point me to a page) how the three Unsharp Mask controls work? I read the GIMP Help, but I'm not sure if the assumptions and procedures for sharpening real world photography hold for our fractals...
- Hal Lane
######################## # hallane@earthlink.net ########################
-----Original Message----- From: Fractint [mailto:fractint-bounces@mailman.xmission.com] On Behalf Of david Sent: Tuesday, May 24, 2016 4:35 AM To: Fractint and General Fractals Discussion <fractint@mailman.xmission.com> Subject: Re: [Fractint] FOTX -- May 22, 2016
I use GIMP all the time, being of sound Linux mind ;). And the Lanczos setting for image resizes.
Try the Unsharp Mask filter: Filters > Enhance > Unsharp Mask. You might like the results.
Thanks for the images, now just need to start rendering them at full Fractint max size, downsample to 25%, and print posters!
On 05/23/2016 04:33 PM, Harold Lane wrote:
The late Paul N. Lee suggested I try the image processing program GIMP. I had not tried GIMP until my anti-aliasing results in Photoshop (using their best down-sampling algorithm -- "Bi-cubic sharper") on this Win 8.1 computer came out poorer quality than the Lanczos algorithm used in my not-supported (on Win 8.1) image processing program I had used on my Win-XP computer: "CompuPic".
GIMP *has* the Lanczos algorithm, and I have used it on this set of my investigations of Jim's FOTX -- May 22, 2016 image. I followed the Lanczos down-sampling with a small amount of sharpening to attempt to remove some of the "softness" of edges created by that algorithm.
I like the results much better.
I'm not yet familiar with GIMP and inadvertently left the default "Progressive" checkbox checked, so the JPG images appear in three "waves" of increasing resolution. I'll try and remember to uncheck that box for the next set of images.
-- David W. Jones gnome@hawaii.rr.com authenticity, honesty, community http://dancingtreefrog.com _______________________________________________ Fractint mailing list Fractint@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/fractint --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
participants (3)
-
david -
Harold Lane -
Jim Muth